Whatever you measure, you can not avoid noise. You can only reduce it, up to a certain point, which is defined by the noise originating in the sample. In most situations, the dominat noise is due to noise contributions from the sensor(s) themselves, the rest of the quipment, plus the preamplifier noise, pluse the noise from the environment are dominant, and those can be worked upon. Such a work can be tedious and lengthy, though. To start with, try to measure with and without the sample; if there is no or little difference, you know that the sample is not to be blamed. Do not exclude the possibility that the noise is due to a failure of an electronic component that needs to be replaced. If you have a possibility to do it (a friend, maybe?) try a comparative measurement with another instrument. These are generic considerations. For anything specific you should describe your exact experimental setup, grounding/screening precautions, and the model of the measuring instrument. Noise can be very specific and sensitive to such details.

What is the frequency of the obscuring signal? Is it fixed? If it is fixed, then you can use a filter circuit to remove it (notch filter is ideal in this case). Also, you have to make sure that the generated signal voltage is high enough and within specifications of your analyzer, otherwise you have to use amplifier/conditioner.

Hello Kaphle,

if you use a signal vibrator it is recommended to use lock in amplifier to measure signal response. You can tune such amplifier in frequency and phase to your applied frequency and therefore exclude noise frequencies like 50/60 Hz or 100/120Hz. A selective amplifier would help too but it is not phase sensitive.

With Regard

R. Mitdank